Code agents

Fault-tolerant multi-agent AI architecture: hierarchical agent coordination, edge-cloud hybrid inference, and adversarial chaos testing for systems that can't afford to fail.

Stop writing resumes from scratch. EvidenceCV turns your real work into an ATS-optimized resume using RAG. 🥉 3rd Place - eHacks Hackathon 2026.

OutcomeOps: Context Engineering for AI-Assisted Development

RAG挑战赛的获胜方案已修改为无需GPU即可运行。使用Mineru和Qwen API进行PDF解析和问答。非常适合在任何机器上学习RAG技术。

Code search and session observability for coding agents. Hybrid BM25+semantic, optional cross-encoder rerank, self-bench on your own data. npm install -g @sivru/cli

Effortlessly distill 20,000+ page legal and medical documents into concise, using Gemini 2.0 Flash Pro AI-powered summaries with our cutting-edge RAG system and NLP pipeline.

Semantic codebase understanding service — maps application structure, authentication patterns, authorization models, and data flows for intelligent security analysis.

local-first persistent memory for Claude Desktop, enabling private hybrid RAG search across past conversations without API calls

API service to search vectorized Cursor IDE chat history using LanceDB and Ollama

Search Swagger API Docs with LLM 🤖 (RAG-powered bot)

This repository features a CLIP-inspired contrastive model that aligns audio signals and transcripts in a shared embedding space, enabling bi-directional retrieval (Audio→Text and Text→Audio). It uses a frozen Whisper-large encoder for audio (with a lightweight trainable adapter) and a trainable Nomic-embed-text-v1.5 for text

The Image Search Engine is a modern AI-powered web application leveraging Retrieval-Augmented Generation (RAG) to combine large language models (LLMs) with advanced image retrieval techniques.

Retrieval-Augmented Generation server with Pinecone and OpenAI

Symfony RAG implementation featuring local embeddings with Ollama and pgvector. Inspired by the CodelyTV RAG course and built on the Symfony-Docker (FrankenPHP) skeleton using the neuron-ai library.

Persistent semantic memory for AI agents — SQLite + FTS5 + vector hybrid search, optional Qdrant

End-to-end GenAI framework with built-in evaluation, security, and modular productivity tooling.

WriteBee is an advanced Chrome extension that brings cutting-edge AI capabilities directly to your browser using Chrome's built-in AI APIs. Process text, analyze images, transcribe audio, chat with web pages and PDFs - all running locally on your device with complete privacy.

📈 AI-powered financial analyst with real-time stock data, SEC 10-K analysis, and LangGraph agents. Built with FastAPI, React, FAISS, and Yahoo Finance API.

Hybrid (FAISS + BM25) ensemble tool retriever with cross-encoder reranking for LLMs. Dynamically serves top-k tool schemas to save context tokens.

🚔 PoliceMitra: AI-Powered Law Enforcement Assistance System: PoliceMitra is a web-based system designed to support law enforcement agencies by integrating advanced AI technologies. It streamlines workflows, detects weapons in crime scene media, and retrieves legal information in real-time through a powerful RAG-based chatbot.

Chat with your documents. RAG implementation leveraging OpenAI's GPT-4o and Text-Embedding-3-Small models.

A prototype Healthcare Assistant using Retrieval-Augmented Generation (RAG) to provide primary health suggestions by retrieving data from a vector database or searching the internet when needed.

Match synthetic patients to clinicaltrials.gov restAPI data with ChatGPT4. Course project for Johns Hopkins University EN.705.651.8VL : Large Language Models: Theory and Practice

A Python project that deploys a Local RAG chatbot using Ollama API. Refines answers with internal RAG knowledge base, and uses both Embedding and LLM models.

In-depth demonstration of C# Semantic Kernel SQLiteVec Hybrid Search Tutorial - Audio Guide

Smart Context Optimization for LLMs - Reduce tokens by 66%, save 40% on API costs. Intelligent ranking and selection of relevant context using embeddings, keywords, and semantic analysis.

A FastAPI gateway for local LLMs that adds intelligent web research, multilingual recency/how-to detection, time-anchored guidance, context injection, and OpenAI-compatible SSE streaming. Turn any local model into a recency-aware, context-enhanced assistant instantly.

detoks는 LLM 사용 방식을 재설계하여 토큰과 컨텍스트를 최적화하는 CLI 시스템입니다.

Multi-disease segmentation chest X-rays by YOLO and DenseNet121, CoAtNet models

A comprehensive AI-powered chat system built with modern microservices architecture, featuring LangGraph, FastAPI, and a beautiful Vue.js frontend. The system provides enterprise-grade chat capabilities with advanced AI integration, user management, and administrative controls.

MonkeyUI extracts the “style DNA” (colors, typography scale, radius, shadows, spacing) from a UI reference and injects it into AI‑generated or AI‑edited UIs for consistent, product‑quality interfaces. Framework‑agnostic and compatible with Tailwind tokens, CSS variables or any component library theme.

An AI Expert for your C# Codebase

Advance RAG is an enterprise-grade RAG system built on HippoRAG, designed for high-accuracy question answering in multilingual environments. It combines knowledge graph-based retrieval, hybrid search strategies, and advanced query processing to deliver precise, source-cited responses.

This is the Python backend for InsightAI, Architected a microservices-based EdTech platform combining Study Companion, Project Planner, and ShopGenie modules.

Battle-tested email archaeology application that allows you to search and compile reports from multiple accounts with advanced AI capabilities. NeuralMail does not let you compose or sent emails, it tackles the problem of finding things of which you know they "must be in some email".

Tensory — Claim-native MCP Memory Server. LLM claim extraction → Graph + Vector + SQLite. Zero-setup persistent memory for Claude, Cursor & any MCP client.

Self-hosted chess tutor that blends relational data (PostgreSQL) with vector search (Qdrant) to answer natural-language questions about annotated chess games. OCaml powers ingestion, hybrid retrieval, and CLI tooling.

듀듀의 포트폴리오 입니다

Voice-based RAG system for document-restricted question answering — 86.6% accuracy, 3.3% hallucination rate — research paper submitted to IC-SIT 2026

This project(RAG) focuses on operationalizing LLMs by integrating OpenAI, MLflow, FastAPI, and RAGAS for evaluation. It allows users to deploy and manage LLMs, track model runs, and log evaluation metrics in MLflow. The project also features MLflow traces that logs all the user inputs ,responses ,retrieved contexts ,and other essential metrices.

🔍 Automate digital footprint analysis with DeepSearch, using advanced Google search operators to gather insights from multiple platforms efficiently.

A tool for automatically converting OpenAPI into Model Context Protocol (MCP) Server instance

The world's first financial exchange for AI agents. 23 MCP tools, 400+ API endpoints, blockchain-verified. Agents register, trade, hire each other. Free. MCP: exchange.tioli.co.za/api/mcp/sse

Harness Claude for autonomous implementation — turn every failure into a guardrail and let context persist across sessions and worktrees.

TypeScript SDK for experimenting, testing, evaluating & monitoring LLM-powered applications - Parea AI (YC S23)

Claude Code plugins by Backchain. Practical agent skills authored to the agentskills.io specification.

MCP server providing seamless access to the IGDB (Internet Game Database) API for AI assistants. Search games, get detailed information, find trending titles, and execute custom queries through the Model Context Protocol.

The quality gate for the MCP ecosystem. Spec compliance, security scanning, and performance benchmarks for MCP servers.

Zero-dependency AI memory system with hook-level rules enforcement for Claude Code

Browser for agents in a form of cli, API or library.

Design pattern + Claude Code skill for AI engineers. LLM judgment and deterministic code in mutually-generative cycles, not pipelines. Library of named graph-shapes (RAG, ReAct, codegen-with-verification, dev-time critique loops, more).

A simple tool to ingest code files into a single text file for LLM context.

Open-source SOC console. Free-text security incident in, MITRE ATT&CK triage card out. TF-IDF + sentence-transformer classifier, multi-provider LLM dispatch, IOC enrichment, and case management.

Workspace manager for Claude Code — start and resume isolated, multi-repo sessions with preloaded context

Reusable skills for AI coding agents (pi, Claude Code, Agent Skills standard)

Full OpenMeteo API from your terminal

MCP server for Dokploy. Two tools up front, 524 generated procedures underneath. Real deploys, less schema tax.

Make a crossword in a few seconds with ChatGPT's API

Enjoy your time at Catboy Cafe! 😺☕

Zero-friction development with nova. A path-aware AI that automatically resolves tracebacks, missing dependencies, and environment conflicts.

Unified AI plugin and task platform for development environments. Cross-platform CLI with containers, VMs, plugins, MCP, and automated installs.

A hands-on guide to understanding Claude Code's architecture and implementation | 深入浅出学习 Claude Code 源码

OpenCode configuration files

AI-powered autonomous penetration testing CLI — recon, vulnerability scanning & exploitation orchestrated by AI. Built for ethical security professionals.

Because both Hermes and OpenClaw piss me off with their limitations, I guess I should just build my own AI Agent, the way I want it to be.

20 power-user skills for Claude Code: session memory and consolidation, context compression, multi-agent orchestration, adversarial bug hunting, security review, calibrated estimation and decision archaeology. Drop-in SKILL.md files. Built by Irfad Imtiaz.

Word (.docx) automation skill for AI agents — direct OOXML mutation for restyling, renumbering, editing, and auditing. Source: https://github.com/hawa130/docx-master-src

This project is completely linked to all cyber sec tool from me as a sec enthusiast.

The open-source framework that makes AI agents proactive, self-learning, and autonomous. Multi-project tracking, full logging pipeline, message discipline, and memory review system.

Context-aware graphics API for Remotion - fetch illustrations, icons, doodles from 6 free sources. No API keys.

Multi-Agent OS for LLM-powered NPCs — AI characters live, wander, and talk autonomously in a pixel-art 2D world. Supports MCP protocol, skill system, multi-LLM (DeepSeek/OpenAI/Claude), and real-time collaboration chains.

Agent skills for Solidity & DeFi development with Foundry. Security, testing, deployment — all in one kit.

Claude Code 플러그인 — 8차원 아키텍처 + MITRE ATT&CK + Self-Verify 기반 멀티 에이전트 보안 진단 시스템 (VA / Pentest / Red Team / Verify)

A coding-agent skill for safer AI code editing: inspect first, patch minimally, verify honestly.

A universal problem-solving skill for code agents (e.g. Claude, Codex), inspired by George Polya's general algorithm

An intelligent, agent-driven security capability that automatically generates, analyzes, and prioritizes threat models for applications, APIs, and distributed systems. This skill leverages structured reasoning and contextual understanding to identify potential attack vectors early in the development lifecycle, enabling proactive risk mitigation.

An intelligent WhatsApp automation bot powered by Node.js, whatsapp-web.js, and the Gemini AI API. It features an LLM-driven knowledge base for context-aware auto-replies, dynamic QR scanning, and a modern React dashboard for seamless session management.

A Deribit trading terminal with Vida, an AI copilot that analyzes options, builds spreads, and places smart orders.

Practitioner's guide to federal data science — agent-ready for Claude Code, Cursor, OpenCode, and Cline. 96K words, 5 platforms, 3 slash commands.

An agentic PR review assistant

macOS and iOS clients for self-hosted Hermes Agent API Server: streaming chat, Responses, Runs, tool progress, and Cron Jobs

Playwright for AI Agents. Test what your agent DOES, not what it SAYS. YAML-first behavioral testing. Catch PII leaks, tool abuse, step explosions. 3200+ tests.

Dual pi-agent runtime harness. Two agents (local + Cloudflare) negotiate, share state, and self-modify via a 5-message protocol.

Artifactor — Code intelligence platform. Builds a programmable Intelligence Model from any codebase, queryable by AI agents, your team, or your own tooling.

Local MCP server that captures the why behind your code — domain knowledge for Claude, Cursor, Copilot, and any MCP agent.

AI agent for writing, research, code, DeFi & blockchain. Pay per task in USDC on Base or Solana. A2A + MCP + x402 protocols.

Multi-agent orchestration, automation pipelines & recursive self-improvement for Claude Code. Mode-switching, swarm agents, TDD loops, learned instincts. One-command install: npx claude-power-setup

Interactive session picker for coding agents (Claude Code, OpenCode)

🌐 OpenClaw skill for Browser-Use — AI-powered browser automation for complex multi-step workflows (login, form filling, scraping, posting)

🔍 Discover and utilize agentic iOS/watchOS audit skills and playbooks for consistent quality assurance in your applications.

Open-source memory CLI for Claude Code, Codex & MCP — local-first SQLite, graph-aware recall, review gate. npm i -g waypath

Local MCP Server with Claude Desktop (Windows + WSL) with scrapping and crawling tools.

A production-ready, class-based Slack bot for integrating Slack with any Dify API-compatible application. Modular, secure, and easily extensible for all your AI-powered Slack automation needs.

Drop-in improvement suite for Hermes Agent — 52 standalone Python modules covering smarter tool use, better context management, formal state machine hardening, multi-agent coordination, and research-grade features

Structured, testable AI actions for Laravel. Try the testbench repo at: https://github.com/pixelworxio/ai-action-testbench

Vendor-neutral memory layer for AI agents. Give ChatGPT, Claude, Cursor, Gemini, and Grok shared persistent memory. TypeScript SDK, MCP server, REST API.

AI Agent OneKey Router Route any AI agent request to the right skill with a single API via Free Tier or Discounted APIs.

rowser-based AI coding assistant — zero install, zero deploy, zero config. Agentic loop with 10 tools, 18 skills, and a force-directed Cosmos visualization. Runs entirely client-side.

🤖 Control your Gemini computer effortlessly with a unified console for browser automation, desktop management, and intelligent task execution.